A transation of Hilary Mason's classifier from her video class, "Machine Learning with Web Data".
Classify.scala is a straight port to Scala. I used the Chalk library where Hilary's original code used nltk, but beyond that I tried to follow the code to see if Scala was faster than Python; it was in my runs.
ClassifyAkka.scala is an attempt to break out parts that could be parallelized. The code to turn a line into features is functional and standalone. In this version I'm having trouble with incrementFeature and incrementCat. I hope to be able to make these methods more functional after I read "Functional Programming in Scala" and put more into the workers.